Skip to main content

Full text of "Observational constraints on modified gravity models and the Poincaré dodecahedral topology"

See other formats

Observational constraints on modified gravity models and the Poincare dodecahedral 


M.C. Bento,i'E| O. Bcrtolami,i'|3 M.J. RebouQas,2^[l and N.M.C. Santos^^l 

^ Departamento de Fisica, Instituto Superior Tecnico, 
Avemda Rovisco Pais, 1049-001 Lisboa, Portugal 
Centra Brasileiro de Pesquisas Fisicas 
Rua Dr. Xavier Sigaud 150 
22290-180 Rio de Janeiro - RJ, Brazil 
^Institut fiir Theoretische Physik, Universitdt Heidelberg 
Philosophenweg 16, 69120 Heidelberg, Germany 
(Dated: February 5, 2008) 

We study observational constraints on models that account for the accelerated expansion of 
the universe via infrared modifications to general relativity, namely the Dvali-Gabadadze-Porrati 
braneworld model as well as the Dvali- Turner and Cardassian models. We find that significant 
constraints can be placed on the parameters of each model using type la supernovae data together 
with the baryon acoustic peak in the large scale correlation function of the Sloan Digital Sky Survey 
of luminous red galaxies and the Cosmic Microwave Background Radiation shift parameter data. 
Moreover, by considering the Poincare dodecahedral space as the circles-in-the-sky observable spa- 
tial topology, we show that the detection of a such a non-trivial topology would provide relevant 
additional constraints, particularly on the curvature parameter, for all models. 

PACS numbers: 98.80.-k, 98.80.Es, 98.80.Jk, 95.36.+X 


Models where gravity is modified by soft very long- 
range corrections, normally inspired in braneworld con- 
structions, are an interesting approach to account for 
the recent accelerated expansion of the universe, with 
no need for dark energy. One of the simplest co- 
variant modified-gravity models is based on the Dvali- 
Gabadadze-Porrati (DGP) braneworld model pj, as gen- 
eralized to cosmology by Deffayet j2|. In this model, 
gravity is altered at large distances by the slow leakage 
of gravity off our 4-dimensional brane universe into the 
5-dimensional bulk spacetime, leading to a modification 
of the Friedmann equation in a cosmological context. At 
small scales, gravity becomes effectively bound to the 
brane and 4D gravity is recovered to a good approxima- 
tion. Crucially for our purposes, it was shown by Deffayet 
that the model exhibits cosmological solutions with a self- 
accelerating phase at late times. An interesting variation 
of this proposal has been suggested by Dvali and Turner 
(hereafter referred to as DT model). Another possibil- 
ity, also originally motivated by extra dimensions physics, 
is the modification of the Friedmann equation by the in- 
troduction of an additional nonlinear term proportional 

'Electronic address: Also at Centro de 

Fi'sica Teorica de Parti'culas, Instituto Superior Tecnico, Avenida 

Rovisco Pais, 1049-001 Lisboa 

TElectronic address: utl.ptl 

"'"Electronic address: reboucas@cbpf. br| 

5 Electronic address: n.santos@thphy| 

to p", the so-called Cardassian model ■ 

We analyze current constraints on the parameters of 
these models, as provided by the so-called gold sam- 
ple of 157 type la supernovae (SNe la) ^j, as well as 
the baryon oscillation acoustic peak (BAO) in the large 
scale correlation function of the Sloan Digital Sky Sur- 
vey (SDSS) 7] and the cosmic microwave background 
radiation (CMBR) shift parameter Q. We consider a 
non-fiat prior. Notice that joint SNe-CMBR-BAO con- 
straints have already been considered for the ACDM and 
quintessence models in Ref. and for an f(R) modified 
gravity model in piil |. While this work was in progress, 
the corresponding analysis for the DGP model has ap- 
peared, see Ref. [llj, where the authors conclude that 
both fiat DGP and ACDM models are within the 1 sigma 
contour, but the latter provides a better fit to the data. 
In what concerns the DT and Cardassian models, con- 
straints from supernovae data alone have previously been 
studied, see |[12j and references therein. 

Likewise for dark energy models, one expects the pa- 
rameters of modified gravity models to be affected by 
the geometry of the universe. The description of the 
universe as a metrical manifold, requires the characteri- 
zation of its geometry and its topology; hence, a key is- 
sue regarding our understanding of the universe concerns 
its 3-dimensional geometry and topology. Studies of the 
CMBR such as the ones performed by the Wilkinson Mi- 
crowave Anisotropy Probe (WMAP) allow for testing ge- 
ometry, which is related with the intrinsic curvature of 

^ Other braneworld models could also be considered, e.g. the 
model proposed in Ref. 5|. 


the 3-dimensional space. On the other hand, topology 
concerns global properties of space such as its shape and 
size and, clearly, 3-geonietry restricts but does not de- 
termine the topology of its spatial section. However, in a 
locally spatially homogeneous and isotropic universe, the 
topology of its spatial sections determines the sign of its 
local curvature and therefore dictates its geometry. 

Different strategies and methodologies have been de- 
vised to probe a putative non-trivial topology of the spa- 
tial sections of the universe (see, e.g. Refs. fT3, Ts^ for 
reviews and details on cosmic crystallographic methods) . 
For instance, the so-called circles-in-the-sky method, is 
based on the presence of multiple images of correlated 
circles in the CMBR maps . In a space with a detect- 
able non-trivial topology, the last scattering sphere (LSS) 
intersects some of its topological images along pairs of cir- 
cles of equal radii, centered at different points on the LSS, 
with the same distribution of temperature fluctuations, 
ST. These pairs of matching circles will be imprinted on 
the CMBR anisotropy sky maps regardless of the back- 
ground geometry or detectable topology 0,01 ■ Hence, 
it follows that in order to probe observationally a non- 
trivial topology, one should examine the full-sky CMBR 
maps in order to extract the correlated circles, and use 
their angular radii and the relative position of their cen- 
ters to probe a putative non-trivial topology of the spatial 
sections of the observable universe. 

In particular, it has been shown that the Poincarc do- 
decahedral space topology (see Section IV. A) accounts 
for the low value of the CMBR quadrupole and octopole 
moments measured by first year WMAP data , which 
has been confirmed by the most recent WMAP data anal- 
ysis 10], and fits the temperature two-point correlation 
function |2ll |23 | . Recently, the Poincare dodecahe- 
dral space |22'|, through the circles-in-the-sky method, 
has been considered as the observable spatial topology 
of the universe in order to reanalyze the current SNe la 
plus X-ray gas mass fraction constraints on the density 
parameters of matter (flm) and dark energy (Qa) in the 
context of the ACDM model with the result that 
it considerably reduces degeneracies. The circles-in-the- 
sky method has also been used to place constraints on 
the parameters of the generalized Chaplygin gas (GCG) 
model mm El, as discussed in Ref. ^J. In that work, 
by using both the Poincare dodecahedral and binary oc- 
tahedral topologies, it has been shown that these spatial 
topologies through circles-in-the-sky could provide addi- 
tional constraints on the As parameter of the GCG model 
as allowed by the SNe la observations. 

Given these encouraging results it is natural to use this 
strategy to constrain the parameters of modified gravity 
models as well. To this end, we will consider the Poincarc 
dodecahedral space topology to reanalyze current con- 
straints on the parameters of the DGP, DT and Cardas- 
sian models, in a joint analysis with the observational 
constraints mentioned above, namely the gold sample of 
SNe la, as well as the baryon oscillation acoustic peak in 
the large scale correlation function of the SDSS and the 

CMBR shift parameter. 


Modified gravity models explore the possibility that 
there is no dark energy, and consider instead that in- 
frared modifications to general relativity exist on very 
large scales, accounting in this way for the observed late 
time acceleration of the universe. 

A. Dvali-Gabadadze-Porrati model 

One of the simplest covariant modified-gravity models 
is based on the DGP braneworld model 1], as generalized 
in Ref. to a FLRW brane in a Minkowski bulk. 

In the DGP model, standard model gauge fields are 
confined to a (3-|-l)D brane residing in an non-compact 
(4+l)D bulk, with different scales of gravity on the brane 
and in the bulk. The gravitational part of the action is 
given by 

S = 


d X dw 


where M5 denotes de 5D Planck mass, Mpi is the 4D 

Planck mass, g'^' is the trace of the 5D metric 
{A,B = 0, 1, 2, 4), w is the extra spatial coordinate, 
g the trace of the 4D metric induced in the brane, 
gf_iv{x) = gllJ{x,w = 0), and where i?5, R4 are the 
5D and 4D scalar curvatures, respectively. This gravita- 
tional action coupled to matter on the brane leads to a 
modified Friedmann equation, which can be written as 





is a length scale beyond which gravity starts to leak out 
into the bulk. 

Rewriting the above equation in dimensionless vari- 
ables = Px/Pcrit with Peril = 3 Mpi Hq/8tt and 
the energy density in the component x today, we get 



where Hq is the Hubble expansion parameter today and 
z is the redshift, and we have taken into account that, 
at present, the universe is matter dominated, hence p ~ 


Pra- Moreover, fJ^ = — t^t^- is the present curvature 
parameter and 



The constraint equation between the various compo- 
nents of energy density at z = is then given by 

= 1 


It has been shown that the observed recent acceleration of 
the universe can be obtained from the extra contribution 
to the Friedmann equation by setting the length scale 
close to the horizon size |28| |29|1 . 

B. Dvali- Turner model 

Inspired in the above construction, Dvali and Turner 
considered a more generic modification of the Friedmann 
equation hereafter referred to as DT model 






Notice that (3 is the only parameter of the model: the 
case /3 — 1 corresponds to the DGP model, /3 — to the 
cosmological constant case, and = 2 to a "renormal- 
ization" of the Friedmann equation. A stringent bound 
follows from requiring that the new term does not inter- 
fere with the formation of large-scale structure, /3 < 1, 
whereas the successful predictions of Big-Bang nucle- 
osynthesis impose a weaker limit on /3, namely, (3 < 1.95. 
Moreover, it can be shown that this correction behaves 
like dark energy in the recent past, with equation of state 
Weff = —1 + /3/2, and w = —1 in the distant future; 
moreover, it can mimic w < —1 without violating the 
weak-energy condition 3]. 

The expression for the Hubble expansion as a function 
of redshift is then 


where flr^ is now generalized to 



which means that the constraint between the various den- 
sities at 2; = is given by 

fi™ + rife + 2Jn7(i - rjfe)'^/2 ^ ^ 


C. Cardassian model 

We will also consider the so called Cardassian 
model 3 , which explains the current acceleration of the 
universe by a modification of the Friedmann equation 
consisting basically in the introduction of an additional 
term proportional to p" 

Stt , , „, k 
— (p-f 6p")- — 




where h and n are constants, and we have added a cur- 
vature term to the original Cardassian model. As in the 
previous cases, in this model the universe is composed 
only of radiation and matter (including baryon and cold 
dark matter) and the energy density required to close the 
universe is much smaller than in standard cosmology, so 
that matter can be sufficient to provide a flat (or close 
to flat) geometry. 

For n < 1 the second term becomes important if 
z < 0(1); thereon it dominates the Friedmann equation 
and yields a oc 

^2/3n fQj. 

ordinary matter, so acceleration 
will occur if n < 2/3. There are two main motivations 
for the introduction of the extra term, namely terms of 
that form typically when the universe is embedded as 
a three-dimensional surface (3-brane) in higher dimen- 
sions _30J or, alternatively, it may appear in a purely 4D 
theory due to an extra contribution to the total energy 
density as would be the case if there were some unknown 
interactions between matter particles 31]. 

In a matter dominated universe, Eq. (|ll|l can be rewrit- 
ten as 

-^) =r!™(i + z)3 + i7fe(i + z)2 

+{i-n,^-nk){i + z) 



Notice that the case n — Q corresponds to the ACDM 

Finally, it is worth pointing out that these models 
have been thoroughly scrutinized from the observational 
point of view using constraints from CMBR, SNe la and 
large scale structure. For recent studies see, for instance, 
Refs. llLJ^,^^^^,^^^^]. 


A. Constraints from SNe la 

For our analysis, we consider the set of SNe la data re- 
cently compiled by Riess et al. ^ known as the gold sam- 
ple. This set contains 157 points: 143 points taken from 
the 230 Tonry et al. ji^l data plus 23 points from Barris 
et al. |41,J and 14 points discovered using HST 0. Vari- 
ous points where the classiflcation of the supernovae was 
uncertain or the photometry was incomplete have been 
discarded, thus increasing the reliability of the sample. 


The data points in the gold sample are given in terms of 
the distance modulus 

fJ-ohsiz) = m{z) - Mohsiz) , 


and the respective errors (T^^i,^ (z), which already take into 
account the effects of peculiar motions. The apparent 
magnitude m is related to the dimensionless luminosity 

Z?i(z) = i^5(y(z)) , 


where S{x) = (sin(a;), sinh(a;), x) for < 0, fi^ > and 
Ofe = 0, respectively, by 

m(z) = M + 5logi„D l{z) 
We have defined a new function 


y{z) = ^TT^I 

The is calculated from 






Mobs(^:») -M' - 5 login -PLth(g»;«») 

1 2 


where M' = A4 — Afobs is a nuisance parameter, are 
the model parameters and Dnhiz; ai) is the theoretical 
prediction for the dimensionless luminosity distance de- 
termined using the modified Friedmann equations. 

B. Constraints from the SDSS baryon acoustic 

In order to further remove degeneracies intrinsic to the 
distance fitting methods, it is interesting to consider also 
the effect of the baryon acoustic peak of the large scale 
correlation function at 100/i~^ Mpc separation, detected 
by the SDSS team using a sample of LRG (Tj] . The posi- 
tion of the acoustic peak is related to the quantity 






S {y{zirg)) 



which takes the value Ao — 0.469 ± 0.017, and where 
Zlrg = 0.35 0. We have neglected the weak dependence 
of Ao on the spectral tilt. The baryon acoustic peak is 
taken into account by adding the term 

2 ^ / 




to the x^, where a a is the error of ^o- 

FIG. 1: A schematic illustration of two antipodal match- 
ing circles in the LSS. These pair of circles come about in 
all globally homogeneous positively curved manifolds with 
a circles-in-the-sky detectable topology. The relation be- 
tween the angular radius 7, angular sides Vinj and Xisa is 
given by the following Napier's rule for spherical triangles, 

cos 7 = tan Tinj cot Xlas ■ 

We should point out that there is a level of uncer- 
tainty in the measurement of A due to uncertainties es- 
sentially on rim (notice that uncertainties on the baryon 
density Sib are constrained by the CMBR and Big Bang 
nucleosynthesis to be smaller than about 2%). Also, one 
should notice that the baryon acoustic oscillations were 
analyzed using a fiducial ACDM model and the full data 
set was compressed to a constraint at a single redshift Q . 
As pointed out by Dick et al |42|, the reduction of the 
data was intended to be valid for the case of a ACDM 
model and robust for models with a constant equation of 
state, but may give rise to significant systematic errors 
for the models we are considering. Although a reanaly- 
sis of the baryon oscillation data in the context of these 
models would no doubt be desirable, one may argue that 
substantial changes are not to be expected given that the 
modifications of gravity we are considering are supposed 
to alter general relativity only at the Gpc scale. The 
same can be said about topology given that it affects 
only the low modes of the CMBR spectrum. 

C. Constraints from the CMBR shift parameter 

It is expected that when the cosmological parameters 
are varied, there is a shift in the whole CMBR angular 
spectrum, that is ^ TZ£, with the shift parameter TZ 








































































TABLE I: Best fit parameters for the ACDM, DGP, DT and Cardassian models for different combinations of observational 
constraints (SN = SNe la gold sample, BAO = SDSS baryon acoustic oscillations, CMBR = CMBR shift parameter and 
T =Poincare dodecahedral space topology for 7 = 50° ±6°). 

being given by 


where zi^s = 1089 18]. The results from CMBR 
(WMAP, CBI, ACBAR) data correspond to 7^n = 
1.716 ± 0.062 (using results from Spergel et al. J^). 
We include the CMBR data in our analysis by adding 

2 I'Uo-n 


to the total function, where TZ is computed for each 
model using Eq. (|20|) . 


In the framework of standard Friedmann-Lemaitre- 
Robertson- Walker (FLRW) cosmology, the universe is 
described by a space-time manifold which is decom- 
posed into A^4 = M X M3 and endowed with a locally 
(spatially) homogeneous and isotropic metric 


1 — kr^ 



where, depending on the sign of the constant spatial 
curvature fc, the geometry of the 3-space M3 is either 
Euclidean (fc = 0), spherical (fc = 1), or hyperbolic 

Thus, since our 3-dimensional space M3 is chosen to be 
one of the following simply-connected spaces, Euclidean 

M."^, spherical S'^, or hyperbolic space H'^, depending on 
the sign of the constant spatial curvature fc, it is a com- 
mon misconception that the Gaussian curvature k of M3 
is all one needs to establish whether the 3~space where 
we live in is finite or not. However, it is known that the 
great majority of constant curvature 3-spaces, M3, are 
multiply-connected quotient manifolds of the form R'^/F, 
S'^/F, and H'^/F, where F is a fixed-point free group of 
isometrics of the corresponding covering space. Thus, for 
example, for the Euclidean geometry besides R'^ there are 
6 classes of topologically distinct compact and orientable 
spaces M3 that can be endowed with this geometry, while 
for both the spherical and hyperbolic geometries there is 
an infinite number of non-homeomorphic (topologically 
inequivalent) manifolds with non-trivial topology that 
admit these geometries. On the other hand, since the 
ultimate spatial topology has not yet been determined 
by cosmological observations, our 3-dimensional space 
may be any of these possible quotient manifolds. 

Quotient manifolds are compact in three independent 
directions, or compact in two or at least one independent 
direction. In compact manifolds, any two given points 
may be joined by more than one geodesic. Since the ra- 
diation emitted by cosmic sources follows geodesies, the 
immediate observational consequence of a nontrivial de- 
tectable spatial non-trivial topology^ of M3 is that there 
will be multiple images of either cosmic objects or specific 
spots on the CMBR. At very large scales, the existence of 
these multiple images (or pattern repetitions) is a physi- 
cal effect that can be used to probe the 3-space topology. 

^ The extent to whi ch a non-trivial topology may be detected was 
discussed in Refs. l43l. 


0.25 0.3 


-0.01 - 
-0.03 - 
-0.07 ^ 
-0.08 - 


FIG. 2: Confidence contours (68.3%, 95.4% and 99.7%) in tiie - ilk plane for tlie ACDM (left) and DGP (riglit) models 
obtained with the SNe la gold sample plus the SDSS acoustic peak data and CMBR shift parameter (dashed lines). Also shown 
are the contours obtained assuming in addition a D space topology with 7 — 50° ± 6° (full). 

In 5D braneworld models, the universe is described by 
a 5-dimensional metrical orbifold (bulk) O5 that is mir- 
ror symmetric (Z2) across the 4D brane (manifold) A44. 
Thus, the bulk can be decomposed as O5 = M4 x Ei ~ 
R X M3 X El, where Ei is a Z2 symmetric Euclidean 
space, and where is endowed with a Robertson- 
Walker metric Eq. (|22|l . which is recovered when w — 
for the extra non-compact dimension. In this way, the 
multiplicity of possible inequivalent topologies of our 3- 
dimensional space, and the physical consequences of a 
non-trivial detectable topology of M3 (possible multiple 
images of discrete cosmic sources, circle-in-the-sky on the 
LSS) is brought on the braneworld scenario. 

A. Poincare Dodecahedral Space Topology 

The Poincare dodecahedral space I? is a 3-manifold of 
the form S^/F in which F = /* is the binary icosahe- 
dral group of order 120. It is represented by a regular 
spherical dodecahedron (12 pentagonal faces) along with 
the identification of the opposite faces after a twist of 36°. 
Such a space is positively curved (k=l, < 0), and tiles 
the 3-sphere S"^ into 120 identical spherical dodecahedra. 

The observed values of the power measured by WMAP 
of the CMBR quadrupole {£ = 2) and octopole {£ = 3) 
moments, and the sign of the curvature density fifc = 
-0.02 ± 0.02 reported by first year WMAP data anal- 
ysis team il8|, which has been reinforced by the three- 
year WMAP observations (cf. Table 11 of Ref. fl^). 
have motivated the suggestion of the Poincare dodecahe- 
dral space topology as an explanation for this observed 
large-angle anomaly in the CMB power spectrum [2^ . 
This observation has sparked the interest in the dodec- 
ahedral space, which has been examined on its various 
features El El El El ■ In particular, it turns out 
that a universe with the Poincare dodecahedral space 

section squares with WMAP data in that it accounts 
for the suppression of power at large angle observed by 
WMAP [3 Ell, and fits the WMAP temperature two- 
point correlation function [Hl^, retaining the standard 
FLRW background for local physics. Notice however that 
a preliminary search of the antipodal matched circles in 
the WMAP sky maps predicted by the Poincare model 
has failed |4J|. A second search of the correlated cir- 
cles for the "D space was not conclusive either 47J . This 
absence of evidence of correlated circles may be due to 
several causes, such as the Sunyaev-Zeldovich effect, lens- 
ing and the finite thickness of the LSS, as well as possible 
systematics in the removal of the foregrounds, which can 
damage the topological circle matching. Thus, it is con- 
ceivable that the correlated circles ma y h ave been over- 
looked in the CMBR sky maps search |23| • 

Regarding the compatibility of the Poincare dodecahe- 
dral space topology with string theory, it has been shown 
that, on account of the Adams-Polchinski-Silverstein con- 
jecture on the instability of non-supersymmetric AdS 
orbifold, the Poincare dodecahedral space topology can- 
not arise as a model for the spatial sections of acceler- 
ating braneworld cosmological models in the framework 
of string theory However, while the braneworld 

paradigm has often been referred to as string-inspired, 
the models we are considering are not committed to that 

B. The Circles-in-the-Sky Method 

An important class of constant curvature positively- 
curved 3-spaces with a non-trivial topology is comprised 
by the globally homogeneous manifolds. These manifolds 
satisfy the topological principle of (global) homogeneity, 
in the sense that all points in M3 are topologically equiv- 
alent. In particular, in these spaces the pairs of matching 


circles will be antipodal, as shown in Figure ^ 

The Poincare dodecahedral space T) is globally homo- 
geneous, and give rise to six pairs of antipodal matched 
circles on the LSS, centered in a symmetrical pattern as 
the centers of the faces of the dodecahedron. Figure Q] 
shows a pair of these antipodal circles. Clearly the dis- 
tance between the centers of each pair of these correlated 
circles is twice the radius r.inj of the sphere inscribed in 

It then follows from the use of trigonometric relations 
(known as Napier's rules) for the right-angled spherical 
triangle shown in Fig. Ogives origin to a relation between 
the angular radius 7, the angular sides and radius 
Xigg of the LSS, namely 

Xiss ^ tan" 

tan r. 

cos 7 


Eq. (|24() makes apparent that Xiss depends on the cosmo- 
logical scenario. Moreover, Eq. H23|l with Xiss given by 
Eq. (|24|l together with the ratio Hq /H for each modified- 
gravity model yield a relation between the angular radius 
7 and the cosmological parameters of each model. Thus, 
they can be used to set bounds (confidence regions) on 
model parameters. To quantify this, we consider a typ- 
ical angular radius 7 — 50° estimated in Ref. |23| for 
the Poincare dodecahedral space and, since the measure- 
ments of the radius 7 do involve observational uncertain- 
ties on the model parameters from the detection of the 
spatial topology, we take into account these uncertain- 
ties; in order to obtain conservative results, we consider 
57 ~ 6°, which is the scale below which the circles are 
blurred for the Poincare dodecahedral space case [20|| . 

where rinj is a topological invariant, equal to tt/IO for 
the the space V, and the distance Xiss to the origin is 
given by 

Xiss = y{ziss) 



0.2 0.4 


a" -0.05 



FIG. 4: Confidence contours (68.3%, 95.4% and 99.7%) for the Cardassian model, in the flm — n (top) and Qk ~ n (bottom) 
planes, obtained with the SN -l-BAO + CMB data (left) and including also a V space topology with 7 — 50° ± 6° (right). Slfc 
or are fixed at the value that minimizes the x^- 

C. Constraints from cosmic topology 

The effect of cosmic topology is taken into account by 
adding a new term to the a-s 






The value of xiss is given by Eq. H24|l and the uncertainty 
'^xiss comes from the uncertainty 6^ of the circles-in-the- 
sky method. The theoretical value of xiss for each model 
is obtained from xiss = y{ziss) combined with the respec- 
tive expansion law. 


We have performed a best fit analysis with the mini- 
mization of the total x^, 

2 2,2 

X ^XSN+Xsds 

X^ , 


for the modified gravity models mentioned above using 
a MINUIT ,4^] based code. Notice that we marginalize 

analytically over A4' and that we allow the parameters 
P and n to vary in the interval ] — 10, 1 ] and ] — 10, 2/3 [ , 
respectively. In the cases where topology is taken into 
account, we use the prior flk < since the Poincare 
dodecahedral space is positively curved. Our results are 
summarized in Tabled 

In Fig. El we show the 1, 2 and 3 cr confidence con- 
tours in the fim — plane for the ACDM (left) and 
DGP (right) models obtained using the SNe la gold sam- 
ple, SDSS acoustic peak and CMBR shift parameter data 
(dashed lines). We also show the contours obtained as- 
suming a V space topology with 7 = 50° ± 6° (full lines). 
This figure together with Table |l] shows that the best- 
fit DGP model is slightly open whereas the the best-fit 
ACDM model is slightly closed. Notice also that the ef- 
fect of the CMBR shift parameter is to push the best 
fit values for Qk towards a flat universe, as first pointed 
out in Ref. j^J, while the SDSS baryon oscillation data 
constrains basically flm, in agreement previous works 
m El H m. Moreover, we see that the effect of in- 
cluding the topology constraint leads to a reduction of 
degeneracies, particularly relevant for the curvature pa- 


rameter, flk- 

Our results for the DT and Cardassian models are 
shown in Figs.|21and01both with and without the topol- 
ogy constraint. The SNe+CMBR+BAO constraints lead 
to best-fit models that are, in both cases, closer to flat 
than the best-fit DGP model and closed spaces are pre- 
ferred. Constraints on Qm are similar for all models. The 
effect of topology is, for these models, again clearly im- 
portant regarding the curvature parameter, flk, and does 
not affect significantly the remaining parameters. 

FIG. 5: Lines of constant angular radius, 7, in the Qk — 
plane for the DGP (solid) and ACDM (dashed) models. 

In Fig. Owe show how our results regarding the DGP 
and ACDM models depend on the angular radius of the 
circles, 7. This figure displays the lines of constant 7 in 
the flk — plane for these models. We see that when 
the angular radius increases Qk becomes more negative 
and that, for any given value of the angular radius, the 
ACDM model prefers a less curved universe as compared 
with the DGP model. It is also clear that the bounds of 
topological origin on flk are expected to be very similar 
for both models, for any given angular radius in the inter- 
val 60° < 7 < 10°; however, for 7 > 70° the distinction 
between these bounds for the ACDM and DGP mod- 
els becomes more important. Finally, this figure shows 
that the detection of circles predicted by the V topology 
restricts substantially the range for ilk , regardless of the 
angular radius 7. Clearly additional limits on this density 
parameter will arise for a specific value for 7 and related 
uncertainty S'y. We find that, although the best fit for flk 
changes for different values of the angular radius, 7, this 
dependence is not very strong (see Fig. O . For instance, 
if we assume a smaller value for 7, e.g. 7 = 11° ± 1°, as 
suggested in Ref. i the allowed regions will be shifted 
slightly towards values of flk closer to 0. On the other 

hand, changes in the uncertainty of the angular radius 
alter the area corresponding to the confidence regions. 

We have analyzed observational constraints on models 
that account for the accelerated expansion of the universe 
that account for the accelerated expansion of the universe 
via long-range corrections to the Friedmann equation, 
namely the Dvali-Gabadadze-Porrati braneworld model 
as well as the Dvali- Turner and Cardassian models. Us- 
ing type la supernovae data together with the baryon 
acoustic peak in the large scale correlation function of 
the Sloan Digital Sky Survey of luminous red galaxies 
and the Cosmic Microwave Background Radiation shift 
parameter data, we find that significant constraints can 
be placed on the parameters of these models. 

In general relativity, as well as in any metrical theory 
of gravitation of some generality and scope, a common 
approach to cosmological modeling commences with a 
space-time manifold endowed with a Lorentzian metric. 
The metrical approach to modeling the physical world 
has often led physicists to restrict their studies to the 
purely geometric features of space-time, either by ignor- 
ing the role of spatial topology or by considering just 
the simply-connected topological alternatives. However, 
since the topological properties of a manifold are more 
fundamental than its metrical features, it is important 
to determine to what extent physical results related to 
a FLRW universe are constrained by the topology of its 
spatial section. 

The so-called circles-in-the-sky method makes appar- 
ent that a non-trivial detectable topology of the spatial 
section of the universe is an observable attribute, and 
can be probed for any locally spatially homogeneous and 
isotropic space. By assuming the Poincare dodecahedral 
space T> as the circles-in-the-sky detected topology of the 
spatial sections of the universe, we have re-analyzed the 
joint SNe-|-CMBR-|-BAO constraints on the abovemen- 
tioned models. The main outcome of our analysis is that 
the detection of a non-trivial spatial topology of the Uni- 
verse through the circles-in-the-sky method would give 
rise to additional constraints on the curvature parameter 
for the models we have considered. 


M.J.R thanks CNPq for the grant under which this 
work was carried out, and is grateful to J.S. Alcaniz for 
useful discussions. The work of M.C.B. and O. B. was 
supported by Fundagao para a Ciencia e a Tecnologia 
(FCT, Portugal) under the grant POCI/FIS/56093/2004. 

[1] G.R. Dvali, G. Gabadadze G, and M. Porrati, Phys. Lett. [2] C. Deffayet, Phys. Lett. B 502 199 (2001); 
B 485 208 (2000). 


[3] G. Dvali and M.S. Turner, astro-ph/0301510 

[4] K. Freese and M. Lewis, Phys. Lett. B 540, 1 (2002). 

[5] V. Sahni and Y. Shtanov, JCAP 0311, 014 (2003). 

[6] A.G. Riess et aL [Supernova Search Team Collaboration], 

Astrophys. J. 607, 665 (2004). 
[7] D. Eisenstein et al, Astrophys. J. 633, 560 (2005). 
[8] J. R. Bond, G. Efstathiou, M. Tegmark, Mon. Not. Roy. 
Astron. Soc, 291, L33 (1997); A. Melchiorri, L. Mersini- 
Houghton, C. J. Odman and M. Trodden, Phys. Rev. D 

68, 043509 (2003). 

[9] K. Ichikawa and T. Takahashi, arXiv:astro-ph/0511821 
[10] M. Amarzguioui, O. Elgaroy, D. F. Mota and T. Multa- 

maki, arXiv:astro-ph/0510519 
[11] R. Maartens and E. Majerotto, astro-ph/0603353' 
[12] M.C. Bento, O.Bertolami, N.M.C. Santos, and A.A.Sen, 

Phys. Rev. D 71, 063501 (2005). 
[13] I.N. Bernshtein and V.F.Shvartsman, Sov. Phys. JETP 
52, 814 (1980). 

[14] M. Lachieze-Rey and J.P. Luminet, Phys. Rep. 254, 135 
(1995); G.D. Starkman, Class. Quantum Grav. 15, 2529 
(1998); J. Levin, Phys. Rep. 365, 251 (2002); M.J. Re- 
bouga s and G.I. Gomero, Braz. J. Phys. 34, 1358 (2004), 
"astro-ph/0402324 M.J. Rebougas, astro-ph/0504365 

[15] R. Lehoucq, M. Lachieze-Rey, and J.P. Luminet, Astron. 
Astrophys. 313, 339 (1996); B.F. Roukema, Class. Quan- 
tum Grav. 15, 2645 (1998); R. Lehoucq, J.P. Luminet, 
and J.P. Uzan, Astron. Astrophys. 344, 735 (1999); H.V. 
Fagundes and E. Gausmann, Phys. Lett. A 238, 235 
(1998); J.P. Uzan, R. Lehoucq and J.P. Luminet, As- 
tron. Astrophys. 351, 766 (1999); G.I. Gomero, M.J. Re- 
bougas, and A.F.F. Teixeira, Int. J. Mod. Phys. D 9, 687 
(2000); R. Lehoucq, J.P. Uzan, and J.P Luminet, Astron. 
Astrophys. 363, 1 (2000); G.I. Gomero, M.J. Rebougas, 
and A.F.F. Teixeira, Phys. Lett. A 275, 355 (2000); 
G.I. Gomero, M.J. Rebougas, and A.F.F. Teixeira, Class. 
Quantum Grav. 18, 1885 (2001); G.I. Gomero, A.F.F. 
Teixeira, M.J. Rebougas and A. Bernui, Int. J. Mod. 
Phys. D 11, 869 (2002); A. Marecki, B. Roukema, and S. 
Bajtlik, Astron. Astrophys. 435, 427 (2005). 

[16] N.J. Cornish, D. Spergel, and G. D. Starkman, Class. 
Quantum Grav. 15, 2657 (1998). 

[17] M.O. Calvao, G.I. Gomero, B. Mota, and M.J. Rebougas, 
Class. Quantum Grav. 22 1991 (2005). 

[18] D.N. Spergel et al., Astrophys. J. Suppl. 148, 175 (2003). 

[19] D.N. Spergel et al., astro-ph/0603449 

[20] R. Aurich, S. Lustig, and F. Steiner, Class. Quantum 
Grav. 22, 2061 (2005). 

[21] R. Aurich, S. Lustig, and F. Steiner, Class. Quantum 
Grav. 22, 3443 (2005). 

[22] J.P. Luminet, J. Weeks, A. Riazuelo, R. Lehoucq, and 
J.P. Uzan, Nature 425, 593 (2003). 

[23] M.J. Rebougas, J.S. Alcaniz, B. Mota, and M. Makler, 
astro-ph/0511007 Astron. Astrophys., in press (2006); 
M.J. Rebougas and J.S. Alcaniz, astro-ph/0603206 Mon. 
Not. Roy. Astron. Soc, in press (2006); M.J. Rebougas 

and J.S. Alcaniz, Braz. J. Phys. 35, 1062 (2005). 
[24] A.Y. Kamenshchik, U. Moschella and V.Pasquier, Phys. 

Lett. B 511, 265 (2001). 
[25] N. Bilic, G.B. Tupper and R.D. VioUier, Phys. Lett. B 

535, 17 (2002). 

[26] M.C. Bento, O. Bertolami and A.A. Sen, Phys. Rev. D 

66, 043507 (2002); M.C. Bento, O. Bertolami and A.A. 

Sen, Phys. Lett. B 575, 172 (2003); M.C. Bento, O. 

Bertolami and A.A. Sen, Phys. Rev. D 70, 083519 (2004). 
[27] M.C. Bento, O. Bertolami, M.J. Rebougas, and P.T. 

Silva, Phys. Rev. D 73, 043504 (2006). 
[28] C. Deffayet, G. DvaU and G. Gabadadze, Phys. Rev. D 

65, 044023 (2002). 
[29] C. Deffayet, S. J. Landau, J. Raux, M. Zaldarriaga and 

P. Astier, Phys. Rev. D 66, 024019 (2002). 
[30] D. J. H. Chung and K. Freese, Phys. Rev. D 61, 023511 

(2000) . 

[31] P. Gondolo and K. Freese, Phys. Rev. D 68, 063509 

(2003) . 

[32] A. A. Sen and S. Sen, Phys. Rev. D 68, 023513 (2003). 
[33] Y. G. Gong and C. K. Duan, Mon. Not. Roy. Astron. 

Soc. 352, 847 (2004). 
[34] Z. H. Zhu, M. K. Fujimoto and X. T. He, Astrophys. J. 

603, 365 (2004). 
[35] J.S. Alcaniz and N. Pires, Phys. Rev. D 70, 047303 

(2004) . 

[36] O. Elgaroy and T. Multamaki, Mon. Not. Roy. Astron. 

Soc. 356, 475 (2005). 

[37] M. Fairbairn and A. Goobar, |astro-ph/0511029| 

[38] Zong-Kuan Guo, Zong-Hong Zhu, J.S. Alcaniz, and 

Yuan-Zhong Zhang, astro-ph/0603632 
[39] U. Alam and V. Sahni, astro-ph/0511473 
[40] J. L. Tonry et al. [Supernova Search Team Collaboration], 

Astrophys. J. 594, 1 (2003). 
[41] B. J. Barris et ai, Astrophys. J. 602, 571 (2004). 
[42] J. Dick, L. Knox and M. Chu, astro-ph/0603247 
[43] G.I. Gomero, M.J. Rebougas, and R. Tavakol, Class. 
Quantum Grav. 18, 4461 (2001); G.I. Gomero, M.J. Re- 
bougas, and R. Tavakol, Class. Quantum Grav. 18, L145 

(2001) ; G.I. Gomero, M.J. Rebougas, and R. Tavakol, 
Int. J. Mod. Phys. A 17, 4261 (2002); J.R. Weeks, R. 
Lehoucq, and J.P. Uzan, Class. Quantum Grav. 20, 1529 
(2003); J.R. Weeks, Mod. Phys. Lett. A 18, 2099 (2003); 
G.I. Gomero and M.J. Rebougas, Phys. Lett. A 311, 319 
(2003); B. Mota, M.J. Rebougas, and R. Tavakol, Class. 
Quantum Grav. 20, 4837 (2003). 

[44] N.J. Cornish, D.N. Spergel, G.D. Starkman, and E. Ko- 

matsu, Phys. Rev. Lett. 92, 201302 (2004). 
[45] B.F. Roukema et al., Astron. Astrophys. 423, 821 (2004). 

[46] J. Gundermann, astro-ph/0503014 

[47] R. Aurich, S. Lustig, and F. Steiner, astro-p h/05l0847| 
[48] B. Mclnnes, Phys. Lett. B 593, 10 (2004 ). 
[49] http: / / / down load/ 1 
/ 2004_source /tar/minuit32_src.tar.gz